Dean Pym
Purpose: This study aimed to characterise the spatial distribution and co-occurrence of wheelchair (WCS), orbital prosthetic (OP), and electronic assistive technology (EAT) service demand across Wales, and to identify areas of significant spatial co-occurrence between ancillary and WCS demand, to inform integrated planning within the All-Wales Artificial Limb and Appliance Service. Methods: Patient-level registration data were obtained for 56,732 service users and geocoded to Lower Super Output Area (LSOA) using residential postcode. Bivariate Global and Local Moran's I assessed spatial co-occurrence between WCS and ancillary service demand, with Jaccard similarity indices quantifying hotspot overlap. Combined service demand was examined using univariate Global and Local Moran's I. Results: 53.8% of EAT users were concurrently registered with WCS, compared with <4% of OP users. Significant global spatial co-occurrence was identified between WCS and EAT demand (Bivariate Moran's I = 0.0384, p < 0.01), but not WCS and OP demand (Bivariate Moran's I = 0.0109, p > 0.05). Bivariate Local Moran's I identified 68 WCS-EAT and 45 WCS-OP co-occurrence hotspots, with modest and limited overlap with WCS-only hotspots respectively (Jaccard = 0.146 and 0.064). Combined service demand hotspots were largely indistinguishable from WCS-only clusters. Conclusions: EAT and WCS services serve overlapping populations, with over half of EAT users also using WCS, providing a rationale for investigating integrated service delivery. No equivalent case exists for WCS-OP integration. Combined demand metrics are an insufficient proxy for co-occurring need when services differ markedly in scale, and independent spatial analysis of each service is essential for equitable specialist service planning.